Bokeh Application

[berli0z]

The next step is adding a Bokeh Application which will enable to visualize multiple users data through an interactive Dashboard.

[berli0z]

The bokeh dashboard will be developed inside the directory bokeh_app, following this structure:

bokeh_app
│   __init__.py
│   main.py    
│
└───scripts
│   │   tab1.py
│   │   tab2.py
│   │   tab3.py
│   │   tab4.py
│   │   ...
│   │   ....

Ideally, we want to have these tabs available:

• Explore: explore, filter, download your data. (probably, trim dates+make_aggregations+download)

• Diet: information on your information diet. (compare sources, top posts, composition of the feed..)

• Compare: compare with a friend. (venn diagram, impressionOrder mean, source count..)

• Research: divide two groups of users or more and make comparisons, linguistic analysis..

• Youtube: get all downloaded videoId data..

The bokeh dashboard could be served remotely with password access or by uploading your own data (as shown here)

TBD:

• main.py serves different tabs
• cleaning.py provides info on missing data
• explore.py returns a table view of the user data with filters and download option
• stats.py returns 4/4 visualizations (time spent, posts vs ads, top posts, top sources)
• "diet" tab: source count in time, average impression order in time
• comparisons between users: venn diagrams, aggregate sums/means in time
• youtube tab
  ...

[berli0z]

In order to run the bokeh dashboard just navigate to the dashboard folder, then:
bokeh serve --show bokeh_app/